Modified Theory of Gravitation. 95 



sufficiently great wave-length, with atomic matter equally par- 

 taking in such motion, could give rise to no observable 

 phenomena ; if any effects are to be made manifest, as a 

 direct consequence of the primary disturbance, these must 

 arise from the motion of matter with respect to the aether. 

 So far as I have been able to see, after long and careful 

 consideration, there would be nothing capable of affecting 

 interference phenomena, no heating effect, and no production 

 of electromagnetic waves, provided only that one condition 

 were realized. That condition is that the positive and nega- 

 tive electrons should have identical accelerations impressed 

 on them by the direct action of the primary disturbance. 

 The discrimination between positive and negative electrons 

 in relation to gravitational agency is a question presenting 

 many aspects for consideration. Some of these are touched 

 upon in Appendices A and B. 



42. Perhaps the most surprising of the tentative numerical 

 values tabulated above, is that suggested as the rate at which 

 energy is being propagated through the rather, per square 

 centimetre of surface normal to the direction of propagation; 

 the energy so propagated in one second exceeding by many 

 million times the Sun's entire store of available heat. The 

 problem has of course been simplified by limiting the primary 

 disturbance to a single progressive wave-train of definite 

 wave-length; but the result would not have been greatly 

 different if a series of wave-lengths had been included as in 

 § 18, and it is of course quite immaterial whether we assume 

 the primary waves to be travelling indiscriminately in all 

 directions, or predominantly or exclusively in a single direc- 

 tion. It must indeed be admitted that the basis of the 

 estimates put forward in § 33 is nothing better than guess- 

 work ; but, after making all allowances for the very great 

 uncertainty attaching to such conjectures, it appears to be an 

 inseparable feature of the theory proposed that energy should 

 be travelling through the rather on a prodigious scale. 



43. The theory of gravitation which forms the subject of 

 this paper has a good deal in common with its predecessors. 

 In the first place, it is made to appear that gravitational 

 attraction is not an essential and inseparable attribute of 

 matter, the Newtonian constant being theoretically sus- 

 ceptible of increase or diminution, or even of entire suppression 

 through a change of external circumstances. In this respect 

 the present theory is comparable with Le Sage's hypothesis 

 of ultramundane corpuscles, as well as with a scheme put 



